Applying statistical principles to the design, analysis, and execution of health research

The CHÉOS team of biostatisticians has broad expertise in designing and analyzing clinical trials, developing clinical prediction models, and applying causal inference methods to observational data.


Statisticians & Methodologists


Service Requests per Year


Publications per Year

About the Program

The Biostatistics Program at CHÉOS provides support to researchers for all statistical needs in the design and analysis of research studies. Our team of biostatisticians and Scientists has broad expertise in designing and analyzing clinical trials, developing clinical prediction models, and applying causal inference methods to observational data. Areas of specific expertise include the design of Bayesian adaptive randomized trials, stepped-wedge trials, and causal inference methods for complex longitudinal and time-to-event data.

Examples of Our Projects

Our team of Biostatisticians apply their expertise to a wide range of fields and health areas.

Design and Analysis of Randomized Controlled Trials

  • TEC4Home is a research program (PI: Kendall Ho) developing and evaluating the use of home tele-monitoring in patients with heart failure transitioning from hospital back to home. CHÉOS Scientist Hubert Wong and Ph.D. student Derek Ouyang developed and utilized novel methodology in the design of the stepped-wedge trial evaluating the effectiveness of home tele-monitoring on subsequent health utilization and patient satisfaction with care. (Link)
  • Optimal duration of risperidone or olanzapine adjunctive therapy to mood stabilizer following remission of a manic episode: A CANMAT randomized double-blind trial (Link)

Please visit our Clinical Trials Program page for more information.

Development of Clinical Prediction Models

  • Development and validation of a prediction rule for early discharge of low-risk emergency department patients with potential ischemic chest pain (Link)

Design and Analysis of Observational Studies

  • The natural history of cartilage damage and osteoarthritis progression on magnetic resonance imaging in a population-based cohort with knee pain (Link)
  • Characterization of myocardial repolarization reserve in adolescent females with anorexia nervosa (Link)

Analysis of Clinical and Administrative Health Databases

  • Attributable length of stay and mortality of major bleeding as a complication of therapeutic anticoagulation in the intensive care unit (Link)
  • Association between the source of infection and hospital mortality in patients who have a septic shock (Link)
  • Intermediate-term results of total ankle replacement and ankle arthrodesis: A COFAS Multicenter Study (Link)
  • Is the road to mental health paved with good incentives? Estimating the population impact of physician incentives on mental health care using linked administrative data (Link)
  • The impact of cost-sharing of prescription drug expenditures on health care utilization by the elderly: Own- and cross-price elasticities (Link)

Methodology Development

Biostatistics Program Head Hubert Wong leads a research program developing a methodology for pragmatic clinical trials. Current work has focused on improving the efficiency of stepped-wedge trials and on Bayesian adaptive design methodology for speeding up the identification of best treatments in routine care.

Scientist Ehsan Karim leads a research program focusing on developing rigorous pharmacoepidemiological methodologies under the causal inference framework, which harnesses the power of data science approaches, including machine learning, artificial intelligence, and big-data analytics. This research aims to develop methods for appropriately analyzing observational large-scale health care databases to compare the real-world effectiveness of different treatment options, both at a single point and over time.

Projects in this area include:

  • The randomization-induced risk of a trial failing to attain its target power: Assessment and mitigation. (Link)
  • Explaining the variation in the attained power of a stepped-wedge trial with unequal cluster sizes. (Link)
  • Improving efficiency in the stepped-wedge trial design via Bayesian modelling with an informative prior for the time effects. (Link)
  • Can We Train Machine Learning Methods to Outperform the High-dimensional Propensity Score Algorithm? (Link)
  • CRTpowerdist: An R package to calculate attained power and construct the power distribution for cross-sectional stepped-wedge and parallel cluster randomized trials. (Link)